On Tue, 2009-03-24 at 17:07 +0000, Mark Ellis wrote:

> > If I plug my phone into my desktop - Rawhide with synce-hal 0.13.1 - and
> > then enable ICS, what happens is quite interesting. Briefly, it gets the
> > correct 192.168.0.2 address (without manual intervention). Then it
> > reverts to 169.254.2.2. In /var/log/messages, there's a message from
> > dhclient saying it bound to 192.168.0.2 and no message below which would
> > seem to contradict that - but ifconfig says 169.254.2.2. I'm not sure
> > exactly what's going on there. I'll re-do the experiment and paste up
> > the relevant section of /var/log/messages later.
> > 
> 
> I'll admit I very rarely use ICS, so I just tried it.
> 
> It doesn't seem to want to do it if the device is plugged in and no data
> connection is active on the phone. However, if I first get the data
> connection up, then trigger ICS, synce-hal drops off, gets restarted by
> the ICS mode device, fires dhcp and gets an address of 192.168.0.102,
> which is stable. I can then turn off ICS, and synce-hal again resets
> with the mode change to the 169.254.2.2 address.
> 
> By all means send logs if yours insn't behaving.

Yeah, that's what I'd expect to happen, but it's not what actually
happens here. I'll send the logs later.

> > Actually, no, it just entirely ignores the phone at present, does
> > nothing with it. It doesn't show up in a left-click on nm-applet. This
> > is expected, I believe.
> > 
> 
> Again, freely admit I've never tried it, but I wouldn't expect it to
> work for me because I have an interface rule to prevent NM messing with
> that interface. That's why I assumed NM would already work with ICS,
> without that type of rule we used to get problems. I guess they changed
> something ?

I don't really know. I just know that, for instance, when I want to
actually use ICS in anger - on my laptop - I have to run it on the phone
end and then manually do 'dhclient eth1' (yeah, eth1 not rndis0, don't
know what's up with that) at a console. NM doesn't seem to do anything
with the device.

This is Fedora. It may be different on Ubuntu, I guess.

> > <dcbw> there's two ways to handle this with NM
> >  1) ignore sync devices
> >  2) handle sync devices, and only accept zeroconf connection types for
> > them
> > 
> > I didn't ask for any clarification beyond that. WDYT?
> > 
> 
> My only real problem with NM doing anything with a sync mode device is
> that as soon as it does it we have to accept that is going to happen,
> and not try to configure the interface. At that point if a user decides
> they don't want NM, we have a much more complex situation.

Yeah, that was sort of my take on it too. For instance, Mandriva doesn't
use NM, so what would happen there. I've always wanted a standard dbus
interface for 'network connection managers' so that you could talk to
networkmanager and drakconnect and wicd and whatever else with the same
signals, but it's never happened. I'll take this to Dan and see what he
says.

> > As I said, at least in current Fedora Rawhide, NM seems mostly to ignore
> > it...
> > 
> That's definitely not what I was expecting......

I guess I'll ask Dan to clarify on this one, whether this is what he's
expecting or not.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org
http://www.happyassassin.net


------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
_______________________________________________
SynCE-Devel mailing list
SynCE-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/synce-devel

Reply via email to